AI
Anthropic's governance trustees gain scrutiny ahead of $2 trillion IPO
An external board of trustees wielding unusual power over the AI company faces investor questions as Anthropic prepares to go public.
Ars Technica
Anthropic's path to a publicly traded company is forcing the AI industry to reckon with an unconventional governance experiment: a small independent trust that controls the majority of the San Francisco firm's board, with no stake in its equity.
The Long-Term Benefit Trust (LTBT) exists to steer Anthropic toward developing AI responsibly for humanity's long-term interests, even as commercial demands mount. The company has signaled that after its stock-market debut—potentially valuing it at $2 trillion—it intends to preserve this control structure.
How the trust wields power#
Anthropic designed the LTBT to appoint or remove the majority of its directors. Currently, it has selected four of the company's seven board members, including Netflix co-founder Reed Hastings and Novartis chief executive Vas Narasimhan. The trust itself comprises up to five members, though only three seats are filled today.
The group's chair is Neil Buddy Shah, who leads the Clinton Health Access Initiative. Former Federal Reserve chair Ben Bernanke and Richard Fontaine, CEO of the Center for a New American Security, round out the active membership. (Mariano-Florentino Cuéllar, a former California Supreme Court justice who initially joined, stepped down after several months to take a role as Anthropic's chief global affairs officer.)
A potential industry template#
Anthropic sees the trust as a blueprint for how AI companies might govern themselves in the future, according to people familiar with the firm's thinking. The comparison drawn is to GAAP accounting standards—initially voluntary guidelines that later became industry-wide norms. Bernanke's appointment in July, sources suggest, reflects an ambition to build a more durable and institutionalized structure.
Operational scrutiny and oversight#
The trustees maintain active oversight through frequent meetings. They receive advance warning of consequential decisions, including the rollout of new AI models, and convene among themselves weekly. They also meet with Anthropic's leadership every one or two weeks, according to people close to the company.
As Anthropic enters the public market, investors will be evaluating whether this governance model can withstand the pressures of shareholder accountability and market competition.

